Attach ON-Marg columns to an administrative boundary layer

Adds the 2021 Ontario Marginalization Index measures to a layer whose features are one of the geographies ON-Marg publishes, matching on the layer's own key column. The layer is returned unchanged apart from the added onmarg_* columns, so an sf object stays an sf object with its geometry intact.

Usage

add_onmarg(
  x,
  geography = NULL,
  scores = TRUE,
  quintiles = TRUE,
  population = FALSE,
  dimensions = names(onmarg_measure_stems()),
  refresh = FALSE
)

Arguments

x: A data frame or sf object carrying the geography's key column - for example the output of retrieve_census() or retrieve_phu().
geography: Character scalar naming the ON-Marg geography, as in retrieve_onmarg(). NULL (the default) detects it from the key columns present in x, and errors if that is ambiguous or absent.
scores: Logical. Attach the four factor-score columns. Default TRUE.
quintiles: Logical. Attach the four quintile columns, where the geography publishes them. Default TRUE.
population: Logical. Attach ON-Marg's own 2021 population count. Default FALSE, since boundary layers usually carry their own.
dimensions: Character vector naming which of the four dimensions to attach: any of "households_dwellings", "material_resources", "age_labourforce", "racialized_nc_pop". Defaults to all four.
refresh: Logical, passed to retrieve_onmarg().

Value

x with the requested onmarg_* columns added. Features with no ON-Marg row get NA; how many that was is reported in a message, because a key mismatch otherwise looks exactly like a successful join.

Details

The ON-Marg name columns (CSDNAME, HU_NAME, and similar) are deliberately not attached: they restate a name the boundary layer already carries, under a name that would collide with it.

Examples

if (FALSE) { # interactive()
phu <- retrieve_phu_pre2025()
add_onmarg(phu, "phu")
}